The development of short wavelength light emitting devices is of great interest in the semiconductor arts. Such short wavelength devices hold the promise of providing increased storage density for optical disks as well as full-color displays and white light sources when used in conjunction with devices that emit light at longer wavelengths.
One promising class of short wavelength light emitting devices is based on group III nitride semiconductors. As used herein, the class of group III nitride semiconductors includes GaN, AIN, InN, BN, AlInN, GaInN, AlGaN, BAIN, BInN, BGaN, and BAlGaInN. To simplify the following discussion, “GaN semiconductors” includes GaN, and group III nitride semiconductors whose primary component is the GaN as in GaInN, AlGaN, BGaN, and BAlGaInN.
Light emitting diodes (LEDs) are fabricated on a GaN semiconductor having an active layer that generates light by recombining holes and electrons. The active layer is sandwiched between p-type and n-type contacts to form a p-n or n-p diode structure. A p-electrode and an n-electrode are used to connect the p-contact and n-contact, respectively, to the power source used to drive the LED. The overall efficiency of the LED may be defined to be the light emitted to the outside generated per watt of drive power. To maximize the light efficiency, both the light generated per watt of drive power in the active layer and the amount of light exiting from the LED in a useful direction must be considered.
A considerable amount of effort has been expended in prior art devices to maximize the light that is generated from the active layer per watt of drive power. It should be noted that the resistance of the p-type nitride semiconductor layer is much more than the resistance of the n-type nitride semiconductor layer. When the p-electrode is formed on the p-type nitride semiconductor layer, a semiconductor junction or ohmic junction is formed. In either case, there is a voltage drop across the junction, and hence, power is wasted at the junction. To reduce this voltage drop , the p-electrode is usually much wider than the n-electrode to lower the contact voltage.
While increasing the size of the p-electrode increases the amount of light generated in the active region per watt of input power, it leads to a decrease in the amount of light that exits the device, since most of the light exiting the device must now pass through the p-electrode. Accordingly, attempts have been made to maximize the transmittance of the p-electrode. A p-electrode having a transmittance of 40 to 50% has been constructed utilizing a multi-layered film of nickel and gold having an 8 nm gold film layer on a 1 nm of nickel layer. However, even with this relatively high transmittance, there is still considerable room for improvement.
In addition, this transparent p-electrode is too thin for bonding to the electrical conductors used to deliver the power to the LED. Hence, a thicker p-electrode region is required to form a bonding pad. A multi-layered film of nickel and gold having a thickness of several hundreds of nanometers is often used as the bonding pad. The bonding pad is typically a rectangle of the order of 100 microns on a side. Hence, a significant amount of light is lost in the thicker bonding pad regions.
However, even with the best prior art designs, the amount of light exiting the LED is 50% of that generated in the active region. If attempts are made to increase the output by using thinner p-electrodes, the resistance of the electrode increases. As a result higher drive voltages are required to overcome the increased resistance, and efficiency drops.
Broadly, it is the object of the present invention to provide an improved LED design.
It is a further object of the present invention to provide an LED with increased light output efficiency.
These and other objects of the present invention will become apparent to those skilled in the art from the following detailed description of the invention and the accompanying drawings.